Cambridge Analytica CEO talks to TechCrunch about Trump, Hillary and the future

Mike Butcher

A few weeks ago I met with, and interviewed, Alexander Nix, the CEO of Cambridge Analytica. His company has been credited with helping Donald Trump win the U.S. presidency. It’s also been associated with many other controversial political campaigns globally and accused by some of aiding the U.K.’s exit from the EU. He addresses all of these subjects in detail (a shorter summary is here).

The interview, which was recorded, was conducted in private at the IT Arena conference in Lviv, Ukraine.

This is the transcript of the interview, which was 50 minutes long:

Mike Butcher (MB):
You think that digital advertising agencies have ‘got it coming’, such as the WPPs of this world. What do you mean by that?

Alexander Nix (Nix):
Actually probably not digital advertising agencies because they're more progressive. I’m really looking it at the old school traditional creative-led agencies. [For example] within WPP obviously Martin's (Sorrell) has made a huge effort to pivot his business. He’s making a huge effort every day, acquiring 40+ companies a year, something like that. But when you look at the traditional approach to advertising, which is fundamentally driven by guesswork, albeit very intuitive and experienced guesswork… The advert I was thinking about is, do you remember the Cadbury's advert of a gorilla playing the drums? I mean who could have known that that was going to be a national success? I mean, you're telling me they went an opinion-surveyed 5000 people and then decided to make a gorilla advert? Of course they didn’t. They just ‘wing’d it’ and it happened to push people's buttons and it was a great success. Well, that sort of advertising is going to be replaced by highly targeted, very personalized advertising, and that has to be data driven. That's not instead of creativity, that's using data to augment creativity. Data first then creativity, it’s linear.

MB:
Looking at the figures on your website you said you drove 1.5% increase in favourability among people who saw your [US election campaign] ads. Not everyone who becomes more favourable after seeing that ad is going to change their vote for instance. Most of them had been planning to vote for Trump already or [the ad] wouldn't make enough difference to stop them voting for Hillary. But it might for influence a very small percentage of the electorate. That might have been enough to swing Michigan but not the whole election.

Nix:
How many states was the election won over? Four? I mean, winning Michigan, Pennsylvania, Wisconsin, Florida… That was pretty much it.

MB:
Because it came down the electoral college that you were targeting?

Nix:
It always comes down to that. It's always the roadmap to 270. That changes every day. You go in to the election: “We're thinking okay these are 12-14 battleground states.” After six weeks you’re like: “Christ. That one's dead. Got to move. How do how do we get there how do we do the sums to get back to where we want to be?” You know, you don't need to and you can't afford to focus on 50 states. You're looking at... you know you're going to win these ones you're never going to win those, so how do you get to where you want to go?

MB:
You said the people who saw Cambridge Analytica adverts… they likelihood of voting for Trump increased by 1 percent. Isn't it that one of the company's claims?

Nix:
Google’s study was based on the impact of our digital campaign. It said [there was an ] 11.3 percent increase in favourability for Trump. An 8.3 percent increase in intent to vote for Trump. These are significant numbers.

MB:
So in 2017 you can claim you have psychological profiles of 220m U.S. citizens based on five thousand separate data sets?

Nix:
It actually works slightly different to that. We went out and we started to rollout a long form quantitive survey to probe's psychographics.

So we had hundreds of thousands of Americans fill out this survey. Completely independent from that we went and collected hundreds and then thousands of data points on every adult 230 million Americans.

MB:
This was publicly available data?

Nix:
This is publicly available data, this is client data, this is an aggregated third-party data. All sorts of data. In fact, we're always acquiring more. Every day we have teams looking for new data sets.

Let's say based on the personality survey that we've identified five personality types, only, for the whole of America. And let's say for each personality type we've got a hundred thousand people of type A and Type B and Type C. Well, we'll look at the hundred thousand type A personalities, and then we'll have a look at the corresponding data points that we have on those hundred thousand people. We’ll have a look at what attributes they have in common and then we'll build a model based on that. So if we identify that all type A personalities drive a lemon yellow car and wear Wellington boots and have a dog and three children and whatever, we can then make a prediction about everyone else in the universe who has a yellow car, a dog, Wellington boots and say well they're very likely to also have a type A personality based on their data.

The best example I can give you [of] building a model [is that] in England we have a stereotype for conservatives, rural conservatives. They wear a barber [jacket], that Nigel Farage type. They wear a Barber and Wellington boots and have a Labrador, and they all went to private school, and they are therefore going to vote Tory. A stereotype but stereotypes are based on something. Well, that was four data points. Actually, it's quite accurate in England. Imagine if you had 40 data points, or 400 and you extrapolated that they drive Land Rover, they like shooting, they both work in merchant banking and so on. You start to build up those data points and you can very accurately say “Well, I don't know this person's party political affiliation, but I do know they have a Barber and a Land Rover and a dog and they like shooting and work in merchant banking, therefore, they're likely to be a Tory.

MB:
So you call this psychographics?

Nix:
Yes.

MB:
But did you use psychographics in the Trump campaign or didn’t you?

Nix:
No, we didn't. We've been absolutely, incredibly clear about this. We used psychographics all through the 2014 midterms. We used psychographics all through the Cruz and Carson primaries. But when we got to Trump's campaign in June 2016, whenever it was, there it was there was five and a half months till the elections. We just didn't have the time to rollout that survey. I mean, Christ, we had to build all the IT, all the infrastructure. There was nothing. There was 30 people on his campaign. Thirty. Even Walker it had 160 (it's probably why he went bust). And he was the first to crash out. So as I've said to other of your [journalist] colleagues, clearly there's psychographic data that's baked-in to legacy models that we built before, because we’re not reinventing the wheel. [We’ve been] using models that are based on models, that are based on models, and we've been building these models for nearly four years. And all of those models had psychographics in them. But did we go out and rollout a long form quantitive psychographics survey specifically for Trump supporters? No. We just didn't have time. We just couldn't do that.

MB:
You say you’ve been building this data since the late 90s. We know Facebook’s terms of service in 2006 were quite different to what they are now. And actually there was quite a lot of data — scrape is probably too blunt a word — but data which you could pull out of Facebook then that you can’t pull out now. Was it the case that you had huge data sets on these people before the door started to close on the [Facebook] terms of service?

Nix:
Well, actually let me correct you. The company started in the early 90s or late 80s. We were a behavioural science company. We didn't pivot into data analytics till 2012. So, all the data that we collected pre-2012, which was done by the British company SBL group, was collected through quantitive and qualitative research on the ground. Our modus operandi was to go and speak to, say, 100,000 people and start to use that to build on models.

MB:
Speak to them how? Via call centre surveys?

Nix:
Depending on the country… I mean, in America, yes, call centres, the Internet where possible, face to face… But in a country like Nigeria you know you just have teams and teams of students going out there knocking on doors.

MB:
So you're doing that in Nigeria?

Nix:
Well, we've been doing that since our first election was 1994 for Mandela/ANC and since then we’ve done multiple elections every year.

MB:
Another claim CA makes is that you raised nearly $27m for Trump from 950,000 email addresses?

Nix:
No those are two separate things. We ran a ‘small-dollar’ fundraising program. So what we did was we used our data to identify core Trump supporters. These are the diehard Trump supporters of which we estimate there are about 37-38 million people in America. And we then targeted them with a with a donor solicitation or small dollars solicitation campaign to ask them to send donations in. And we built all the data and all the mechanisms to do that. We raised that $27m within the first month of starting work. In total, obviously, that program went on to raise hundreds and millions of dollars.

MB:
Would you consider that to be being pivotal for their campaign?

Nix:
Well I think I think it was extremely pivotal because when Trump won the nomination he had very, very little money. And although he talked about putting a bit of money in himself, and he did put some money and some in as cash, most of it as loans, that's my understanding, you know, you were competing against the machine and she had dollars coming out of everywhere.

Also there was a huge “Never Trump” faction in America. Most of the Republicans didn't support him, and even those who ended up working for him didn't support him. So the RNC, the Republican National Committee, didn't support him. Ultimately they pivoted. A lot of the key RNC members were part of the “Never Trump” faction. They were behind his back. They were trying to destroy him. And eventually they all they did a complete U-turn. Part of the reason why we were thrust into such a prominent role in this campaign is because none of the vendors would support him.

[Being among] Republican vender's is an incredibly hostile environment. They were looking at this candidate and they said “Well, first and foremost we don't like him.” A typical presidential campaign will probably have five or eight different companies support it. You'll have a pollster, you'll have a digital agency, you'll have a TV agency, you'll have a research firm. And then the campaign manager and the campaign committee will choose the best pollsters and the best thing or their best friend or however they figure these things out.

Well, Trump won the nomination and all the Republicans said “Well, he's going to get murdered by Hillary. If we work for him, the establishment of the RNC is going to hate us. We'll never get another dollar in U.S. politics again. So we'll make a quick buck today, but it's going to kill our career tomorrow.” So a lot of them are like “we'd never want to touch this.” So rather than having multiple vendors servicing his campaign, as is traditional, as Hillary had, we walked in there and said “We'll do your data analytics.” And they were like: “There's no one doing research.” [We said] we will do your research. “There's no doing digital” We will do digital. “There's no one doing TV.” “We'll do your TV.” We’ll do your donations. And so overnight it went from being originally just data, to end to end.

MB:
Did you believe you were betting farm, as it were, on the campaign?

Nix:
Look, from my perspective it was an easy bet to make. It was a win-win. I couldn't see the downside. I thought even if Trump didn't prevail, if he didn't win the election.
Look, we're a British firm that was trying to break into the most competitive political market in the world. And you know, we had some mixed press. But what really irritates me is when journalists go to get a quote about our work. And someone says “We worked with Cambridge Analytica and their work didn't really provide anything. It was rubbish.” And then you have a look at who the quote was from and it’s from a direct competitor!

And this is what the journalists haven't quite figured out. A lot of the people that they speak to are people whose lunch we're eating. We walked into [the US] market. We're competing with all the data teams. We’re competing with all the digital teams, all the TV teams, all the research teams. You’ve seen House of Cards. It's like that. It's the most vicious aggressive political culture both at the candidate level, that Trump is now finding out. At the campaign manager, GC level and at the vendor level. It’s a bloodbath. The knives… [are out]. In DC, everyone’s fucking everyone else.

You see three quotes they're all from people whose business you've stolen. They're saying things like we came across “Cambridge Analytica. It’s All snake oil” [and it’s from] our biggest rival.

MB:
Do you think there's a mischaracterization of the tools used by targeted advertising campaigns, or so-called ‘custom audience’ campaigns, as being described as “dark advertising” campaigns?

Nix:
There's no question that the marketing and advertising world is ahead of the political marketing the political communications world. And there are some things that I would definitely [say] I’m very proud of that we're doing which are innovative. And there are some things which is best practice digital advertising, best practice communications which we’re taking from the commercial world and are bringing into politics.

Advertising agencies are using some of these techniques on a national scale. For us it's been very refreshing, really breaking into the commercial and brand space… walking into a campaign where you're basically trying to educate the market on stuff they simply don't understand. You walk into a sophisticated brand or into an advertising agency, and the conversation [is sophisticated] You go straight down to: “Ah, so you're doing a programmatic campaign, you can augment that with some linear optimized data… they understand it.” They know it's their world, and now it comes down to the nuances. “So what exactly are you doing that's going to be a bit more effective and give us an extra 3 percent or 4 percent there.” It's a delight. You know these are professionals who really get this world and that's where we want to be operating.

MB:
Do you regret the way your own business is being presented in the media?

Nix:
If there's any testament to what's driving the media just have a glance at Hillary Clinton's recent book. The liberal press are supporting their candidate. They got fairly beaten and they're lashing out and trying to destroy every single person and every company that contributed to that defeat. Hillary simply cannot come to terms with it. She's a woman in denial. The liberal press [characterised Cambridge Analytica] as “witchcraft, they treat it is “voodoo” and now it's Russia's fault! They just cannot accept the fact that Hillary was such an unpopular, such a divisive candidate. She failed to mobilize her base and people didn't fundamentally trust her. Rather than looking in the mirror, they much prefer to beat up Cambridge [Analytica] beat up Trump, beat up anyone else. Anything but accept the fact that their candidate wasn't what the people wanted to vote for.

MB:
Would you would use your own methods to improve your own public image?A marketing campaign for yourself to create a better image for your own company?

Nix:
Would we rollout a national behavioural micro-targeting program to promote Cambridge Analytica? It probably wouldn't really achieve what we're trying to do. I mean actually I'm just 'dotting the I's and crossing the T's' on a book which talks about our methodology and our approach to communications and I was speaking to the publisher about whether we should might target that. It’s a book hasn't been published yet, it’ll come out next month, just talking really in quite technical terms about how communication is changing, what how technology is impacting that, what data is doing to advertising and political campaigns and then using a lot of case studies with a lot of real examples of artwork and targeting and psychographics and so forth to illustrate them. We were toying with the idea of [a campaign]. But I think it's too complex to try and use our techniques to promote a book. It was more of a thought exercise right. It would mean having multiple versions multiple titles multiple dust sheets.

MB:
A/B testing for a book? Is the book coming out in the New Year?

Nix:
It’s coming out in Germany first because it was a German publisher that approached us. It will be coming out in about the next month or so and then later in the U.K.

MB:
It is going to be about the company or methodology. Or your world view?

Nix:
The English title is not confirmed yet but I think it will be something like “Mad men to maths men”. It's going to be [about] the evolution of the advertising industry, how data, how psychology, how digital is changing an industry that really hasn't changed very much. And what that disruption means in terms of an industry of you know multi-billion dollar industry and illustrating different campaigns that we've done to show the effectiveness of that.

MB:
On Brexit there were headlines such as ‘The great Brexit robbery, our democracy was hijacked’. How do you react to those?

Nix:
Well, look, I mean you're implying therefore that we were involved in Brexit and part of that “robbery”. We've been, again, crystal clear to all media including The Guardian, who really propagated this story from day one, based on nothing. Carole Cadwalladr has made it her personal mission to come after us again, living in denial about the outcome of the election. She cannot accept that the British people wanted to leave Europe and she's made it her mission to vilify us. We did not work on Brexit. We didn't do a little bit of work. We didn't do a lot of work. We did no work on Brexit. We were not involved in the campaign for either Leave.eu or Vote Leave, at all. And we have been crystal clear on this which is why we're going to be taking them to court and we're going to settle.

MB:
Why do you think you became associated with it then?

Nix:
Well because before the campaigns were launched we were approached by a number of different campaigns, pro and against, to discuss whether there might be a role for us on Brexit. And we had a number of discussions. Obviously, these discussions made their way into the public forum but we meet with hundreds of companies every year and talk about business opportunities. That doesn't mean you engage with them doesn't mean you contract with them and it certainly doesn't mean that you work for them. But The Guardian came out with one or two data points and then created an entire narrative around that that was pure fiction. That went viral and then every other newspaper [piled into it]. If you tell a lie often enough it becomes truth. But even after we came out and denied that again and again and again they just kept propagating the same message.

MB:
But you are working in other campaigns. There's been some controversy about Kenya and South Africa for instance. How do you react to those?

Nix:
Well, you are talking about Bell Pottinger, [which has] paid the consequences for some bad decisions.

MB:
Did you work on those campaigns?

Nix:
No, we weren't involved with the actors. We weren't involved with Bell Pottinger, we weren’t involved with the Guptas. In fact, I know very little about that. And I really can't make a commentary on what happened. I know only what I've read in the paper.

MB:
You worked on the Kenya campaign?

Nix:
We've worked all across Africa.

MB:
Kenya?

Nix:
Well, let's wait till the election's over because we never talk about elections that live, generally, as a rule of thumb. But I can tell you that we worked in Kenya in 2013 on the last election for Kenyatta. That's well documented. I can speak to that. Look, Kenyatta won by 12 points and there were some few irregularities, from what I’ve read, in the way that the election oversight committee… [granted] some of the tenders [for] election equipment. The opposition used these irregularities to challenge the outcome of the election. I think had this been a “Florida” years ago when it came down to half a percent, ok. But when you're talking about a 12 percent victory I think that the court's decision to hold this [new] election was a dreadful decision. I think that it is going to result in dreadful bloodshed, horrific violence. If Kenyatta for any reason doesn't win this election then his supporters are going to feel robbed. And if Odinga’s people don’t, they're going to feel that he's cheated again because that's the perception that Odinga is put out into the public domain.

I can't see this ending well. I think just for the sake of Kenya, for peace in the region, I think it's a dreadful decision. I actually can't even see this election being resolved in the next month. I think it's going to drag on. So, I think the court's decision was shortsighted.

MB:
What's your answer to critics who claimed that you’ve worked on behalf of the Russian government or third party actors connect to them? Either for specifically in relation to Trump's campaign or to other campaigns.

Nix:
We’ve never been asked directly. No one of authority has levelled any direct criticisms to us and, certainly, no one has suggested that we'd been or alleged that we’d been involved as far as I'm aware. We’re not under investigation by anyone. We are helping wherever we can with the understandings of the campaign, like everyone else in the campaign, but there's no investigation into Cambridge [Analytica].

MB:
Isn't the U.S. Congress investigating you in connection with Russian attempts to interfere with the election?

Nix:
No, it's not. The US Congress is undertaking…

MB:
The Atlantic magazine reported it.

Nix:
[Scoffs] Oh then it must be true! I don't know… I mean that's exactly what I'm talking about. My understanding is that the U.S. Congress is undertaking an investigation into Russian interference into the election and they've asked all sorts of people for help into that. That's not suggesting in any way, any way at all, that Cambridge is under investigation. And you know we're more than happy to help. We never worked in Russia. We never worked for Russia. I want to be careful, but I don't think we have any Russian employees in our company whatsoever. We just don't have business in Russia. We have no involvement with Russia, never have done.

MB:
What about for third parties associated with Putin's government?

Nix:
I wouldn't even know who they were and where to begin. I mean, we worked directly for the campaign, as in directly, and we worked for a Super Pac in support of the campaign [called] “defeat crooked Hillary” was it’s unofficial title or “keep the promise” or something.

MB:
You have an investor, Robert Mercer. What sort of independence that give you? He has known political views. Do you feel independent of an investor like that?

Nix:
Well, actually, I'm not going to speak about any of our investors or board members at all because we don't. But I can answer your question which is ‘is our political ideology influenced by other people in the company’ at whatever level? And the answer to that [is this]… We undertake 7 to 9 elections a year, somewhere in the world, for Prime Minister or President. And for as many of those are on the Left of Centre or the Right of Centre. In fact, if you were to total them up, I would probably say - and this is based on a guesstimate - that we’d done more Left-leaning than Right-leaning. Now, clearly in America…

MB:
Can you give me any examples of those Left-leaning campaigns?

Nix:
You’d have to go on our website. I’m sure you could find 25, and you can just see which parties. It’ll all be there. But in America you have to pick a side. You can't flip-flop. You are not encouraged work for the Democrats in one cycle and then move to the Republicans [on the next]. And the reason the Republicans were attractive to us was because the Democrats were significantly leading the tech arms race. Under Obama through Civas and Blue Labs they had pioneered the use of big data. They were using very sophisticated digital technologies. And the Republicans had been left behind. By the time Romney lost in 2012 there was a vacuum. There just wasn't the tech talent on the Right to be able to compete. It was like taking a knife to a gunfight. And so that was the commercial opportunity. Now, the reason for that is because most of the tech community, and I'm going to generalize here, but a lot of the tech community that were politically oriented tended to lean Left. People in Berkeley and MIT, and so forth, would, if they were politically motivated, support Hillary or Obama. Whereas, people who might be more Right of centre, obviously, would look typically to go and work in banking or investment management with those skills. And so because of the dearth of talent the Republicans were getting murdered in the tech arms race. That was the commercial opportunity, that was the one we sought to address. Right? Had it been the other way around it might have been a different story.

MB:
What are your own personal political views? Do you talk about those?
Nix:
We leave our personal ideologies at the door. We think that being “foreign” and objective is an asset in elections. I think that a lot of political campaigns, and in the U.S., a lot of campaign staff and vendors get blinded by their own ideology. They blindly believe in their candidate to an extent that they actually can't see objectively what's happening in the campaign. They can only see what they and all their friends believe and therefore they assume that's representative of 200 million voters take

MB:
[You’re saying] They project themselves onto campaigns?

Nix:
Yeah naturally. And there's something wonderful about coming into a foreign country as an outsider and looking with completely fresh eyes at a political landscape and be able to not have a clouded judgment. And that's what we bring. And equally, it's like a good lawyer representing his or her client. You can't go in there with a preconception of guilty or not. You have to go in there and look at the facts. And that's what we try to do.

MB:
Do you have a blacklist of anyone you wouldn't work for?

Nix:
Oh yes for sure. We only work for mainstream political parties. Tories, Labour, Republicans, Democrats. We steer clear of fringe political parties or minority groups. We're not trying to orchestrate a revolution. We're trying to provide the best tech… communication technology to political parties. As I said before, elections are about 20 percent of our revenue as a company. We’ve got about 20/25 percent in defence and homeland security. The rest is in the brand and commercial space. We are not a political company. We're a tech company, and we see ourselves as tech company. We have a tech culture, where we attract academics. You know a third of our staff are PhDs. We're geeks! Fundamentally, we're a bunch of geeky people who are trying to solve problems and I would say that politics is probably the least desirable division in the company, because it can be divisive and people don't necessarily like to get involved. But, if you give a data scientist a really challenging problem like identifying the ideology of a nation or an issue, a model or something, it's about solving the problem. It is not about trying to promote their own personal ideology or agenda or anything like that.

MB:
You're working with Palantir?

Nix:
No. Palantir was established about nine or 10 years ago now. They're very active of course in the defence, homeland security. They were a pioneer and a leader in this field. Peter Thiel, in applying both his platforms, Gotham and Metropolis, and other work which is slightly more off the radar, it is truly revolutionary. I mean these guys were genuine first movers. And I think that companies like us have caught up. But kudos to them. They were really very early on to the scene.

MB:
I think you count the Pentagon as a client don’t you?

Nix:
We formed a defence/government defence division in 2005. Over that period we've worked for global militaries all over the world. We train a lot of armies in something called PSYOP, which is psychological operations or information operations, which is, sort of communication warfare. It’s trying to understand how to persuade troops not to fight or to persuade your troops to fight. It’s trying to combat hostile behaviour or how do you counter radicalization or counter-terrorism, and so forth. We do a lot of information operation programs ourselves [where] we go in to countries and conduct the research and the campaigns to change behaviour to reduce conflict. And our clients do include, in the U.K, the MOD and the FCO, and in the United States, all the ‘coms’, so NorthCom, Safcom, State Department, Pentagon and various ‘three letter agencies’ and so forth.

MB:
What's your what's your vision for the future? You’ve talked about psychographic profiling, analytics about the amount of data sets there are, and the data you can pull out social media. And of course, we all know that Alexa’s in their houses recording what they are saying. Where do you see things heading? Firstly, which direction do you feel your own company is going in, given the amount of data out there? Secondly, do you think you might end up butting up against the Google and Amazons in data collection?

Nix:
There are several questions there. Let's start with the bigger picture. There are two technologies that I'm really excited by and that everyone is talking about. Clearly, IOT is going to totally radicalize the data market. I think the data business is doomed, myself. It’s a very, very high volume and very low margin [business]. As the internet of things grows, as we have sensors on everything: cars, fridges, TVs then data is going to become ubiquitous. Therefore the volume of data will increase, the price of data is going to go down. You won't be able to sell data in the way you can today. I think people are going to start taking control of their data much more. There is going to be more reciprocity in the way that people share their data with companies like my own and other marketing agencies. But generally, the increase in data is one factor that is going to make analytics companies like ours more valuable. More data is going to need more analytics, period. And then at the other end — if you see it as a sandwich and we're in the middle — you've got blockchain. And by having distributed ledger technology you're going to an ability to have transparency, and to have accountability as to how data is and data sets are being used and implemented, forevermore, in perpetuity. And so, yes the data landscape is getting more frightening with IOT. But on the other hand, it's going to be to be more self-regulated through the Blockchain and it's going to be more transparent. And both of those things are the bread rolls with the analytics being the chess in the middle of the sandwich. Analytics are going to play more and more of a function in deciphering huge quantities of data, making sense of it, applying it into many different areas and then using blockchain technologies to securitise that.

In terms of the advertising industry: Look, I've been very vocal about this but I don't think that, again, I’m sort of some sort of ‘soothsayer’. I think a lot of people in the industry recognize this… And even if they are only whispering to each other… I think the advertising industry is like lemmings on the edge of a cliff. They can't go backwards and forwards looks terrifying. Omnicom, Dentsu, WPP… they're trying to pivot. They're acquiring companies left right and centre. But you can't just buy a data company and squash it together with an ad agency and hope it's just going to work. It doesn't work like that, that it's a different culture. It takes integration. You need to grow these things together. You've got all the consultancies which saw a commercial opportunity in the last two or three years and started acquiring data companies. They're trying to acquire advertising agencies in order to get into this space. Then nipping at the heels of the big conglomerates and taking considerable market share from them. You've got the big brands themselves understanding their data is so valuable they no longer want to give that away to advertising agencies. They can bring some of these capabilities in-house and have their own data analytics and marketing agencies in-house. That's damaging them.

Then you've got the small disruptors like us who are only going to become more numerous. We're eating a piece of the pie. So, I think there is going to be a reckoning, and it's happening now. WPP’s market share has taken a nosedive in the last month or three. But they're not going to be the only ones. It’s going to continue. I'm not suggesting that they're all going to go bust, but I am saying it's “adapt or die”. This market is fundamentally shifting and about time too! This is overdue. Gone are the days where an advertiser does an advert. And as long as the client's wife likes it there are no metrics other than audience recall to quantify its success. Or very few. And they don't care anymore. They can just squeeze a lemon and get some money out of advertisers. I think that advertisers want more accountability, they want more measurement of effectiveness, they want empirical data to be able to justify these enormous multi-billion dollar spends.

MB:
Let’s fast forward say,10-20 years. There will be people who can afford to buy privacy and then there will be plenty of people who will give away their data in return for services, as they do now. Do we do we think that that's a good situation to be in? What's your view?

Nix:
I think gone are the days where people just click a box without really thinking it through, and all their data has just been siphoned away and gone. Grabbed. I think people are recognizing that data is valuable AND that they're [also] saying, “Well actually it's not that sensitive. I don't really care much if people understand my shopping habits or what car I drive. This isn't health or financial data. But it is valuable and why should I just give it away so that other firms and advertisers can make money out of it?” So I see more of a reciprocity. I see people having something like a virtual data wallet. They're going to have control of their data. You'll be able to say to them “Hey can I licence your data or use your data for a certain campaign or purpose?” And there's going to be an exchange there. They're going to say “Yes you can but it's going to cost you 10 percent off that” or “I want this in return.” And I see that market emerging as people take control. I think that's really sensible. I think it gives people more control and therefore it gives them more protection. I think advertisers will need to be a little more targeted in the way that they use data and they gather data, which is probably good. But actually, it's not going to dent the industry or the direction that it's growing in. So I really hope that people like the ICO [Information Commissioner’s Office in the UK which rules on data privacy for individuals], who are suddenly trying to catch up on the last five years of data. I hope that they understand that the regulation needs to be there to protect consumers, but not to stump the growth of an industry that could really do a lot of good. Not just in communications, [but] across all corporate sectors.

MB:
If consumers had more control over their data wouldn’t that start to slightly stump the work that you do?

Nix:
No, I don't think it will because I think things are changing. I think that the older generation, our parents or even older… you know, this is a terrifying brave new world for them. “What do you mean someone knows what car I drive?!” “Yes, mother they do know what car you drive” “But that's awful!” “Why is it awful?” “Well because I didn't want them to know that!” “Do you care?” “That's not the point.” “Well, it is the point!” Actually, the new generation, the next generation, younger than us, they don't care. They actually just don’t care.

MB:
They are used to this world?

Nix:
They are used to this world and they realize it. “Do I care if people know what car I drive, what cereal I eat for breakfast?” And why should they, really? They don't care if they put a picture of themselves blind drunk on Facebook doing something. They don't care. Let alone [someone knowing] what car they drive.

MB:
Do they care if they feel that it might influence an election?

Nix:
Well, that's a good question. Ultimately, I don't think people are naive, now, especially not the Internet generation, the Millennials. I think they do understand what's going on. There's been so much press. It's not about hoodwinking people. Remember, it's the same for all sides. It's not like Trump had some secret sauce that he was employing with Cambridge that the Democrats didn't. Hillary's data and digital teams were up to 200 people or something! Huge! Huge! This was tried and tested. It was a machine! They were doing everything! But the reason they're not in the spotlight is twofold. A: Hillary lost. And B: Trump’s, you know, is somewhat perceived by many as more of a polarizing character. That's why. It's not what we did. I mean, I think if we’d done exactly the same work, no different, but done it for Hillary…

MB:
You think the result would have been the same?

Nix:
No, I think no one would care. The Guardian wouldn't be writing these stupid headlines and nor would The New York Times. They wouldn't care. This is not about Cambridge. It's not about our tech. It's about Trump.

By using Yahoo you agree that Yahoo and partners may use Cookies for personalisation and other purposes